Enabling Product Development and other Life Cycle Processes with PDM PDM: An Enterprise Investment Research Methodology Product Data Management entails the management and classification of product data and the management of change to this information.1 PDMtype applications are greatly impacting the way companies do business throughout their product lifecycle – it is not just CAD drawings anymore. Nine sites representing six different companies participated in the research. For each company, up to six interviews were done: one regarding the site, one to three pertaining to specific programs (legacy to conceptual), and others as appropriate. Data were collected on 24 programs. Over 100 questions were asked covering topics such as requirements, schedule, management support, and training. Just this year, the aerospace industry is expected to spend $10.4 billion on PDM/PLM technologies.2 That much and more in savings has been promised by vendors. Many companies are deriving some benefits of PDM but only after much heartache and hard lessons learned. Also, the full potential of PDM is not being realized as suppliers struggle to catch up. Although many consultants and vendors have provided assistance to industry, they have not focused on the state of implementation and needs across the industry. As Product Lifecycle Management evolves, it becomes imperative for companies to make the most of their investment, taking the opportunity to explore every process throughout the lifecycle and make it work for the enterprise. Enterprise Take-Aways • PDM remains focused on the design stage • Suppliers are moving up the food chain, yet they are behind the curve on product data management capability • Change management and data migration are the biggest challenges/ pitfalls • Lean principles and practices should be used when implementing PDM capability • PDM enables Lean Enterprise Transformation – opportunity to address enterprise value stream • Common reasons for a suboptimal or failed solution may include: lack of management support, continuing a parallel paper process, not compelling users to adopt the system. Two sites were also used for case studies, looking at the front-end process of selecting a PDM, and the tension between change in the organization and evolution of the technology. Results of those will be available postplenary. Range of implementation Number of Employee at each Site periods: 1999 – present 4 3 1 0 0 - 1999 2000-4999 5000-7999 8000-10,999 Implementation Pitfalls 80% non-CM 60% CM 40% PDM+ 20% PDM 2D A M D et 3 a D D at Sc En EB a an g OM St D No ru ra te c w s To t A ing ol na s in ly g si M s od el C s Pr M AE N ocu BO on re M -c m on en f Te Da t st ta D PM ata Fi D el at M dD a ai a O nt D ta th a Sc er D ta he at du a le s 0% C Data Elements • Number of sites that made their planned schedule? • Number of sites with successful first-round training? • Number of sites with consensus on good mgmt support? • Number of programs continuing with a parallel paper process? • Percentage of programs still replicating data? Evolution of Data Management Where Data Elements Are Managed (Future) Data Elements 1. 2. http://www.pdteurope.com/what_is_pdt.html http://www.daratech.com/press/releases/2005/050228.html Functionality 5 4 3 2 1 I n al B PD LC Integration C tio su Co lla bo ra Vi gm Ch an ge M ct ur St ru Pr od w kf lo t e t M gm e as le Re 0% Results from Survey of PDM Maturity (Current) A H 0% c PDM g 20% Sites were asked to rate their PDM implementation in the following areas: 1. Integration of Product Data Across the Product Lifecycle 2. Extent of Supplier/Partner Integration 3. Management of Workflow Electronically Throughout the Product Lifecycle 4. Integration/Compatibility with Current Systems/Applications PDM 20% Do PDM+ 40% Required SW & HW 32% Training and Other 5% Data Quality, Migration… 9% Maturity Assessment Survey PDM+ tin CM 60% ul 60% 40% nonPDM Va non-CM Percentage (Programs) 80% CA D CA 2 M D D et 3 a D Da Sc E EB ta an ng O St D No M ru ra te To ct A win s ol na gs in ly g si M s od el s C Pr M A No ocu BO E n- re M co m nf en Te Da t st ta PM Dat Fi D a el a M d D ta ai a O nt D ta th a Sc er D ta he at du a le s Percentage (Programs) 100% Where Money is Being Spent for PDM Implementations Process Consulting Development 19% 35% Expertise of Interviewees: Lean Change Agents, Six Sigma Blackbelts, Directors of IT and Engineering, Program and IPT Managers, CM, Senior, And Process Engineers ≥ 6 of 21 90% Current Functional Use of PDM (2004) 100% 80% 1 out of 8 1 out of 8 4 out of 8 Why? What are the common mistakes are people making? How should we plan differently? W or C AD Percentage (Programs) 100% 11,000+ Employees PDM Where Data Elements Are Managed (Past) Total implementation experience: 4 Vendor products 34+ years of ‘next-gen’ 2 G D F E Sup/Part Integration Electronic Workflow System Integration A through H represent the sites Possible answers ranged from 1 to 5 For more information, contact Erisa K. Hines erisak@mit.edu 617-258-7984 Research Team: Tom Shields, shields@mit.edu; Jayakanth “JK” Srinivasan,